资源描述
THE 2028 GLOBAL INTELLIGENCE CRISISA Thought Exercise in Financial History,from the FutureWhat if our AI bullishness continues to be right.and what if thats actually bearish?What follows is a scenario,not a prediction.This isnt bear porn or AI doomer fan-fiction.The sole intent of this piece is modeling a scenario thats been relativelyunderexplored.Our friend Alap Shah posed the question,and together webrainstormed the answer.We wrote this part,and hes written two others you can findhere.Hopefully,reading this leaves you more prepared for potential left tail risks as AImakes the economy increasingly weird.This is the CitriniResearch Macro Memo from June 2028,detailing the progressionand fallout of the Global Intelligence Crisis.PrefaceCitriniResearchFebruary 22nd,2026 June 30th,2028The unemployment rate printed 10.2%this morning,a 0.3%upside surprise.Themarket sold off 2%on the number,bringing the cumulative drawdown in the S&P to38%from its October 2026 highs.Traders have grown numb.Six months ago,a print like this would have triggered acircuit breaker.Two years.Thats all it took to get from“contained”and“sector-specific”to aneconomy that no longer resembles the one any of us grew up in.This quarters macromemo is our attempt to reconstruct the sequence-a post-mortem on the pre-crisiseconomy.The euphoria was palpable.By October 2026,the S&P 500 flirted with 8000,theNasdaq broke above 30k.The initial wave of layoffs due to human obsolescence beganin early 2026,and they did exactly what layoffs are supposed to.Margins expanded,earnings beat,stocks rallied.Record-setting corporate profits were funneled rightback into AI compute.The headline numbers were still great.Nominal GDP repeatedly printed mid-to-highsingle-digit annualized growth.Productivity was booming.Real output per hour roseat rates not seen since the 1950s,driven by AI agents that dont sleep,take sick days orrequire health insurance.The owners of compute saw their wealth explode as labor costs vanished.Meanwhile,real wage growth collapsed.Despite the administrations repeated boasts of recordproductivity,white-collar workers lost jobs to machines and were forced into lower-paying roles.When cracks began appearing in the consumer economy,economic punditspopularized the phrase“Ghost GDP“:output that shows up in the national accountsMacro MemoThe Consequences of Abundant Intelligencebut never circulates through the real economy.In every way AI was exceeding expectations,and the market was AI.The only problemtheeconomy was not.It should have been clear all along that a single GPU cluster in North Dakotagenerating the output previously attributed to 10,000 white-collar workers in midtownManhattan is more economic pandemic than economic panacea.The velocity ofmoney flatlined.The human-centric consumer economy,70%of GDP at the time,withered.We probably could have figured this out sooner if we just asked how muchmoney machines spend on discretionary goods.(Hint:its zero.)AI capabilities improved,companies needed fewer workers,white collar layoffsincreased,displaced workers spent less,margin pressure pushed firms to invest morein AI,AI capabilities improvedIt was a negative feedback loop with no natural brake.The human intelligencedisplacement spiral.White-collar workers saw their earnings power(and,rationally,their spending)structurally impaired.Their incomes were the bedrock of the$13trillion mortgage market-forcing underwriters to reassess whether prime mortgagesare still money good.Seventeen years without a real default cycle had left privates bloated with PE-backedsoftware deals that assumed ARR would remain recurring.The first wave of defaultsdue to AI disruption in mid-2027 challenged that assumption.This would have been manageable if the disruption remained contained to software,but it didnt.By the end of 2027,it threatened every business model predicated onintermediation.Swaths of companies built on monetizing friction for humansdisintegrated.The system turned out to be one long daisy chain of correlated bets on white-collarproductivity growth.The November 2027 crash only served to accelerate all of thenegative feedback loops already in place.Weve been waiting for“bad news is good news”for almost a year now.Thegovernment is starting to consider proposals,but public faith in the ability of thegovernment to stage any sort of rescue has dwindled.Policy response has alwayslagged economic reality,but lack of a comprehensive plan is now threatening toaccelerate a deflationary spiral.In late 2025,agentic coding tools took a step function jump in capability.A competent developer working with Claude Code or Codex could now replicate thecore functionality of a mid-market SaaS product in weeks.Not perfectly or with everyedge case handled,but well enough that the CIO reviewing a$500k annual renewalstarted asking the question“what if we just built this ourselves?”Fiscal years mostly line up with calendar years,so 2026 enterprise spend had been setin Q4 2025,when“agentic AI”was still a buzzword.The mid-year review was the firsttime procurement teams were making decisions with visibility into what these systemscould actually do.Some watched their own internal teams spin up prototypesreplicating six-figure SaaS contracts in weeks.That summer,we spoke with a procurement manager at a Fortune 500.He told usabout one of his budget negotiations.The salesperson had expected to run the sameplaybook as last year:a 5%annual price increase,the standard“your team depends onus”pitch.The procurement manager told him hed been in conversations with OpenAIabout having their“forward deployed engineers”use AI tools to replace the vendorentirely.They renewed at a 30%discount.That was a good outcome,he said.The“long-tail of SaaS”,like M,Zapier and Asana,had it much worse.Investors were prepared-expectant,even-that the long tail would be hit hard.Theymay have made up a third of spending for the typical enterprise stack,but they wereobviously exposed.The systems of record,however,were supposed to be safe fromdisruption.It wasnt until ServiceNows Q3 26 report that the mechanism of reflexivity becameclearer.SERVICENOW NET NEW ACV GROWTH DECELERATES TO 14%FROM 23%;ANNOUNCES 15%WORKFORCE REDUCTION AND STRUCTURAL EFFICIENCYPROGRAM;SHARES FALL 18%|Bloomberg,October 2026How It StartedSaaS wasnt“dead”.There was still a cost-benefit-analysis to running and supportingin-house builds.But in-house was an option,and that factored into pricingnegotiations.Perhaps more importantly,the competitive landscape had changed.AIhad made it easier to develop and ship new features,so differentiation collapsed.Incumbents were in a race to the bottom on pricing-a knife-fight with both eachother and with the new crop of upstart challengers that popped up.Emboldened by theleap in agentic coding capabilities and with no legacy cost structure to protect,theseaggressively took share.The interconnected nature of these systems werent fully appreciated until this print,either.ServiceNow sold seats.When Fortune 500 clients cut 15%of their workforce,they cancelled 15%of their licenses.The same AI-driven headcount reductions thatwere boosting margins at their customers were mechanically destroying their ownrevenue base.The company that sold workflow automation was being disrupted by better workflowautomation,and its response was to cut headcount and use the savings to fund the verytechnology disrupting it.What else were they supposed to do?Sit still and die slower?The companies mostthreatened by AI became AIs most aggressive adopters.This sounds obvious in hindsight,but it really wasnt at the time(at least to me).Thehistorical disruption model said incumbents resist new technology,they lose share tonimble entrants and die slowly.Thats what happened to Kodak,to Blockbuster,toBlackBerry.What happened in 2026 was different;the incumbents didnt resistbecause they couldnt afford to.With stocks down 40-60%and boards demanding answers,the AI-threatenedcompanies did the only thing they could.Cut headcount,redeploy the savings into AItools,use those tools to maintain output with lower costs.Each companys individual response was rational.The collective result wascatastrophic.Every dollar saved on headcount flowed into AI capability that made thenext round of job cuts possible.Software was only the opening act.What investors missed while they debated whetherSaaS multiples had bottomed was that the reflexive loop had already escaped thesoftware sector.The same logic that justified ServiceNow cutting headcount applied toevery company with a white-collar cost structure.By early 2027,LLM usage had become default.People were using AI agents who didnteven know what an AI agent was,in the same way people who never learned what“cloud computing”was used streaming services.They thought of it the same way theythought of autocomplete or spell-check-a thing their phone just did now.Qwens open-source agentic shopper was the catalyst for AI handling consumerdecisions.Within weeks,every major AI assistant had integrated some agenticcommerce feature.Distilled models meant these agents could run on phones andlaptops,not just cloud instances,reducing the marginal cost of inference significantly.The part that should have unsettled investors more than it did was that these agentsdidnt wait to be asked.They ran in the background according to the userspreferences.Commerce stopped being a series of discrete human decisions andbecame a continuous optimization process,running 24/7 on behalf of every connectedconsumer.By March 2027,the median individual in the United States was consuming400,000 tokens per day-10 x since the end of 2026.The next link in the chain was already breaking.Intermediation.Over the past fifty years,the U.S.economy built a giant rent-extraction layer on top ofhuman limitations:things take time,patience runs out,brand familiarity substitutesfor diligence,and most people are willing to accept a bad price to avoid more clicks.Trillions of dollars of enterprise value depended on those constraints persisting.It started out simple enough.Agents removed friction.Subscriptions and memberships that passively renewed despite months of disuse.Introductory pricing that sneakily doubled after the trial period.Each one wasrebranded as a hostage situation that agents could negotiate.The average customerlifetime value,the metric the entire subscription economy was built on,distinctlydeclined.When Friction Went to ZeroConsumer agents began to change how nearly all consumer transactions worked.Humans dont really have the time to price-match across five competing platformsbefore buying a box of protein bars.Machines do.Travel booking platforms were an early casualty,because they were the simplest.ByQ4 2026,our agents could assemble a complete itinerary(flights,hotels,groundtransport,loyalty optimization,budget constraints,refunds)faster and cheaper thanany platform.Insurance renewals,where the entire renewal model depended on policyholder inertia,were reformed.Agents that re-shop your coverage annually dismantled the 15-20%ofpremiums that insurers earned from passive renewals.Financial advice.Tax prep.Routine legal work.Any category where the serviceproviders value proposition was ultimately“I will navigate complexity that you findtedious”was disrupted,as the agents found nothing tedious.Even places we thought insulated by the value of human relationships proved fragile.Real estate,where buyers had tolerated 5-6%commissions for decades because ofinformation asymmetry between agent and consumer,crumbled once AI agentsequipped with MLS access and decades of transaction data could replicate theknowledge base instantly.A sell-side piece from March 2027 titled it“agent on agentviolence”.The median buy-side commission in major metros had compressed from 2.5-3%to under 1%,and a growing share of transactions were closing with no humanagent on the buy side at all.We had overestimated the value of“human relationships”.Turns out that a lot of whatpeople called relationships was simply friction with a friendly face.That was just the start of the disruption for the intermediation layer.Successfulcompanies had spent billions to effectively exploit quirks of consumer behavior andhuman psychology that didnt matter anymore.Machines optimizing for price and fit do not care about your favorite app or thewebsites youve been habitually opening for the last four years,nor feel the pull of awell-designed checkout experience.They dont get tired and accept the easiest optionor default to“I always just order from here”.That destroyed a particular kind of moat:habitual intermediation.DoorDash(DASH US)was the poster child.Coding agents had collapsed the barrier to entry for launching a delivery app.Acompetent developer could deploy a functional competitor in weeks,and dozens did,enticing drivers away from DoorDash and Uber Eats by passing 90-95%of the deliveryfee through to the driver.Multi-app dashboards let gig workers track incoming jobsfrom twenty or thirty platforms at once,eliminating the lock-in that the incumbentsdepended on.The market fragmented overnight and margins compressed to nearlynothing.Agents accelerated both sides of the destruction.They enabled the competitors andthen they used them.The DoorDash moat was literally“youre hungry,youre lazy,thisis the app on your home screen.”An agent doesnt have a home screen.It checksDoorDash,Uber Eats,the restaurants own site,and twenty new vibe-codedalternatives so it can pick the lowest fee and fastest delivery every time.Habitual app loyalty,the entire basis of the business model,simply didnt exist for amachine.This was oddly poetic,as perhaps the only example in this entire saga of agents doinga favor for the soon-to-be-displaced white collar workers.When they ended up asdelivery drivers,at least half their earnings werent going to Uber and DoorDash.Ofcourse,this favor from technology didnt last for long as autonomous vehiclesproliferated.Once agents controlled the transaction,they went looking for bigger paperclips.There was only so much price-matching and aggregating to do.The biggest way torepeatedly save the user money(especially when agents started transacting amongthemselves)was to eliminate fees.In machine-to-machine commerce,the 2-3%cardinterchange rate became an obvious target.Agents went looking for faster and cheaper options than cards.Most settled on usingstablecoins via Solana or Ethereum L2s,where settlement was near-instant and thetransaction cost was measured in fractions of a penny.MASTERCARD Q1 2027:NET REVENUES+6%Y/Y;PURCHASE VOLUMEGROWTH SLOWS TO+3.4%Y/Y FROM+5.9%PRIOR QUARTER;MANAGEMENTNOTES“AGENT-LED PRICE OPTIMIZATION”AND“PRESSURE INDISCRETIONARY CATEGORIES”|Bloomberg,April 29 2027Mastercards Q1 2027 report was the point of no return.Agentic commerce went frombeing a product story to a plumbing story.MA dropped 9%the following day.Visa didtoo,but pared losses after analysts pointed out its stronger position
展开阅读全文
浙ICP备2021020529号-1 | 浙B2-20240490 
